A Generic Graph-Based Method for Flexible Aspect-Opinion Analysis of Complex Product Customer Feedback

Abstract

Product design experts depend on online customer reviews as a source of insight to improve product design. Previous works used aspect-based sentiment analysis to extract insight from product reviews. However, their approaches for requirements elicitation are less flexible than traditional tools such as interviews and surveys. They require costly data labeling or pre-labeled datasets, lack domain knowledge integration, and focus more on sentiment classification than flexible aspect-opinion analysis. Related works lack effective mechanisms for probing the customer feedback of complex configurable products. This study proposes a generic graph-based opinion mining and analysis method for product design improvement. First, a customer feedback data preprocessing and annotation pipeline that can incorporate designer-specified domain knowledge is proposed. Second, an intuitive opinion-aware labeled property graph data model is designed to ingest preprocessed feedback data and perform ad hoc opinion analysis. Applying the generic model to a real-world dataset demonstrates superior functionality and flexibility compared to related works. A wider range of analyses is supported in a single model without repeating data preprocessing and modeling. Specifically, the proposed method supports regular and comparative aspect-opinion analysis, aspect satisfaction/influence ranking, opinion trend extraction, and targeted aspect-opinion summarization.

Contextual Meaning-Based Approach to Fine-Grained Online Product Review Analysis for Product Design

“…Identifying the targets of opinion that customers intend to discuss and show main interest in within reviews is crucial to clearly understand customer needs [15], [31]. Therefore, many product design studies have been conducted to identify the opinion targets from online product reviews.…”

“…Existing studies primarily extracted opinion targets at the keyword-, phrase-, and latent topic-levels. Keyword-level identification focused on extracting words related to the product from reviews [11], [12], [13], [15], [16], [17], [20], [22]. Whereas this keyword-level identification represented the most fine-grained analysis in previous studies, the analysis ended at identifying the words related to the product feature terms.…”

A Systematic Procedure for Utilization of Product Usage Information in Product Development

“…In this regard, many recent works have discussed data-driven design [5,26]. Regarding usage data, a set of works have presented the utilization of PUI for product development, for example, as shown in [50][51][52]. These works are quite impressive, but they are rather specific for their tasks.…”

Achieving eco-innovative smart glass design with the integration of opinion mining, QFD and TRIZ

“…In addition, traditional research approaches in sustainable design often overlook the evolving customer preferences from product reviews, as it is challenging for them to leverage the vast volume of unstructured data available on the digital platforms 13 . Incorporating customer feedback into product design is crucial for gaining insights into user experiences and expectations 14 .…”